Cardiac Subcommittee Minutes 02.12.21

Cardiac Anesthesia Subcommittee Minutes

February 12, 2021

12:00pm – 1:00pm EST

Zoom

Abernathy, Jake (Johns Hopkins)

Kertai, Miklos (Vanderbilt)

Andrawes, Michael (MGH)

Kileny, Joel (St. Joseph Mercy Ann Arbor)

Billing, Josh (Vanderbilt)

Krall, Thomas (TJ) (UCSF)

Bailey, Meridith (MPOG)

Lacca, Tory (MPOG)

Bottiger, Bandi (Duke)

Low, Ying (Dartmouth)

Buehler, Kate (MPOG)

Luria, Brent (NYU Langone)

Burrage, Peter (Darmouth)

Mamoun, Negmeldeed (Duke)

Clark, Cantwell (Dartmouth)

Mathis, Mike (MPOG)

Chen, Yunwei (Washington University)

Muehlschlegel, Danny (Brigham and Women’s)

Davies, Eric (Henry Ford Health System)

Neuburger, Peter (NYU Langone)

Douin, Josh (University of Colorado)

Podgoreanu, Mihai (Duke)

Dubovoy, Anna (Michigan Medicine)

Rangrass, Govind (University of Chicago)

Fisher, Clark (Yale)

Reidy, Andrea (Washington University)

Ian Gannon, Michigan Medicine

Rhee, Amanda (Mount Sinai)

Michael Grant, Johns Hopkins

Sera, Valerie (OHSU)

Grewal, Ashanpreet (Ashan) (University of Maryland)

Schonberger, Rob (Yale)

Guruswamy, Jayakar (Jay) (Henry Ford Health System)

Shah, Nirav (MPOG)

Hasmi, Nazish (Duke)

Shook, Doug (Brigham and Women’s)

Janda, Allison (MPOG)

Szymanski, Brooke (MPOG)

Jelacic, Srdjan (University of Washington)

Varelmann, Dirk (Brigham and Women’s)

JohnBull, Eric (Duke)

Vega, Eleanor (Duke)

Katta, Gaurav (Henry Ford Health System)

1. Introductions and interim updates

a. Introductions MPOG Team

i. Allison Janda, MD – MPOG Cardiac Subcommittee Lead

ii. Nirav Shah, MD – MPOG Director of Quality

iii. Michael Mathis, MD – MPOG Director of Research

iv. Kate Buehler, MSN – Clinical Program Manager

b. Cardiac subcommittee is using Basecamp to communicate between meetings: please

accept your Basecamp invitations (sent via email) to stay in the loop

c. Individual Performance E-mail link to new QI Reporting tool

2. Recap from Meetings #1 & 2

a. Call for Measure Survey Results Overview

i. 16 providers completed the survey

ii. Highest rated measures (no overwhelming consensus)

1. #1: Post-bypass hypothermia avoidance (62% listed in the top 3)

2. #2: Glucose management (56% listed in the top 3)

3. #3: Postoperative AKI avoidance (44% listed in the top 3)

4. #4: Hypotension avoidance (44% listed in the top 3)

5. #5: Antibiotic timing (38% listed in the top 3)

iii. MPOG data capture - measure limitations

1. 4 Hours before Anesthesia Start 6 hours after Anesthesia End

b. Overview of variation for 2019 cardiac data:

i. >35.5vC, mean: 85%, SD: 13%

ii. >36.0 C, mean: 66%, SD: 21%

iii. >36.5 C, mean: 35%, SD: 19%

3. Post-Bypass Hypothermia Avoidance

a. Current TEMP-03 Measure:

i. % of patients, with procedures >60 minutes under GA/neuraxial, with at least

one body temperature ≥ 36

ii. Excludes cardiac surgeries

b. Literature Review (see slides for more information)

i. Current guidelines and consensus statements:

1. 2020 Updates from the Adult Cardiac Anesthesiology Section of STS

2. Guidelines for perioperative care in cardiac surgery: enhanced recovery

after surgery recommendations

ii. Current guidelines and consensus statements:

1. STS Practice Guidelines for temperature management while on bypass

2. ERAS cardiac recommendations

3. Current guidelines and consensus statements

iii. Additional Literature

1. Retrospective observational study in cardiac patients

2. Retrospective study evaluating the impact of hypothermia on morbidity

and mortality after off-pump CAB surgery

3. RCT in non-cardiac surgery patients with cardiac comorbidities

4. Systematic review of the impact of mild perioperative hypothermia on

blood loss for non-cardiac Surgery

c. Hypothermia Avoidance Measure

i. Considerations:

1. Threshold: (≥35.5C)

a. Further Discussion?

i. Muehlschlegel: Where are we getting the

measurement? We need to reference where we get the

measurement (i.e. core vs. peripheral). Looking at the

post bypass period, the general group thought that core

was most ideal. We need to do more digging to make

sure we’re noting the source..

ii. Vega: Bladder temperature, if it’s not being linked to

bladder and there is no urine output, it may not be

accurate. Mathis: There will be a bioinformatics

discussion. Temperature data will only be as good as

we can remove artifacts. We should think about edge

cases but during this meeting, we should focus on the

clinical aspects.

iii. Allison: We will dive into the most common sources

once we develop the first draft of the query

specification and dig into the data. It will be a heavy

programmer lift to look into the data on the fringe cases

or those cases with a non-core temp documented to

see if we should look back 5-10 minutes for example.

As far as the actual threshold, are folks okay with this

temperature (>/=35.5)?

iv. Answer: Yes.

2. Timing: Last non-artifact temperature documented, if more than one,

preferentially use core temperature

a. Basecamp confirmed we will look at last non artifact

temperature documents, if more than one, preferentially core

temperature, discussion?

i. Fisher: if not more than one at the last but within the

last 10-minutes may be the best. Mathis: There will be a

look-back period to ensure we are getting the most

accurate temperature. We will only figure this out

when we get out hands on the data.

ii. Consensus: “Last” recorded temperature with

preference of core temperature measures, will review

the data to see if we should include a look-back period,

and if so, what duration, once we dig into the output

from the first draft of the specification.

3. Exclusions: Items to consider: Limit to open cardiac cases; limit to

bypass cases; exclude cases requiring circulatory arrest in light of

hypothermia contributing to neuroprotection

a. Discussion: Muehlschlegel: What’s our outcome? Janda: In the

measure generation itself, we do not have a specific outcome.

The initial purpose is to establish a threshold and after we get

more data we can see how we can use them in research to

possibly look at outcomes such as AKI or mortality.

Muehlschlegel: Depending on the outcome, it would dictate

how I would create the measure. If you’re just being descriptive

it may be valuable to include everything. I would be in favor of

limiting it to cardiac cases. Shook: I agree it should be broad but

if we are using it as a quality measure we want to be certain

about what we’re telling people they want to achieve. It may

not be related to core temperature. We would want

information on everything at first. If we are trying to figure out

all subsets, why can’t we have all three? Janda: I agree, we can

include all in the measure. The providers will have their own

case list and could look specifically at those flagged cases and

determine if there needs to be change in practice. Shook: 35.5 is

a low threshold. Make more inclusive for including off pump

bypass. Douin: I like the idea of categorizing it separately. My

other thought is emergent cases, those are a different subset of

cases and may strongly influence temperature management.

Janda: Do we want to exclude them or include emergency cases

and flag them? Mathis: emergency cases are going to matter

the most. Those are the cases where we can make a difference

vs. straight forward CABG which may not matter that much.

b. Janda: Regarding picking very specific types of cases (HCA or

ECMO) are at the mercy of the quality of the data and how they

are binned. We also have a predicted CPT that has a complex

algorithm and some of the fringe cases could slip through. Part

of the process will be to validate and review the cases.

Abernathy: Is this just temperature at the end of the case? With

what clarification? Why exclude any case? Is there ever a case

that hypothermia leaving the OR is desired? If there isn’t that

case, I would argue to include all cases. Mathis: One thing to

consider is to be able to compare apples to apples, which not

the sole goal of the measure. Are you an outlier because you

have a certain cohort of patients that are more challenging? Any

exclusion criteria we apply can be turned off for research. For

getting some sense of where we stand it will be some effort and

to have truncating at the edges is helpful. Janda: One fringe

case is TVAR where the surgeon refuses to turn on the bair

hugger and that will be determined by practice patterns. When

we get into other measure development, such as glucose

management, we can hone into those, but we are intentionally

leaving this patient at 35.5. Varelmann: Hypothermia leads to

SSI and other complications. Janda: Agreed, some of the

literature focused on surgical site infection and bleeding, which

implies the existence of an incision/sternotomy/larger surgery.

That may be a good reason to limit the group to open cardiac

cases since the temperature management could matter more in

those patients as opposed to a 2 hour TAVR. What is most

helpful for you and your team? Mathis: One other consideration

is that this is just a starting point. What do we want to see as

the first number? If you are interested in a particular sub-

population, you can add filters. Shook: If you can do it for all

the cases and I can easily filter, that is great. If you are going to

do it one way across institutions, I would like to see the first

cases to be pump cases. For my institutions, I would want to see

cases in all ways so I can first see how we are doing at my

institution. Fisher: If we are looking for comparison across

hospitals, I would weed out hypothermic cases because if your

site does not do many, you may think you’re in line with other

sites. Abernathy: If the goal is to leave the room warm and

measure the temperature of the room then it doesn’t matter. I

agree this should be limited to pump cases. Muehlschlegel: The

temperature drop after the case can be dramatic. Janda:

Agreed, I would favor including these HCA cases so when that

case is flagged, it draws attention to these cases. If the case is

never included, people may not know they’re missing an

opportunity to improve. Chen: If we are just including "pump"

cases, are we including ECMO cases? Mathis via chat: Yunwei --

this will be determined by the MPOG 'cardiac case' phenotype --

on MPOG website --> Tools --> Phenotype browser... the

detailed logic (which we can revise/improve) is there --would

love to hear your feedback on it. Janda: Not every site

contributes surgical CPTs. Surgical CPTs are collected over time

and we might not have those to include in time to be added to

the quality measure, and to rule in/our an ECMO case, we don’t

have the granularity in anesthesia CPTs that would be needed to

identify those cases. Mathis: Version 1 of the measure is that

we will have to make sure we add all the nuances of cardiac

cases and we will make adjustments as we go. Muehlschlegel: I

agree, let’s make a decision and review the data first. This will

help us to make sure we agree with the data coming out and

make modifications as we go.

c. Janda: We will come up with a version of this measure based on

these discussions. We may not have a completed consensus but

we do have a good starting point. We will be sending out via

Basecamp or Google Doc.

4. Attributions: Options include: 1) Attending physician signed in for the

most number of minutes; 2) attending physician signed in for the most

number of minutes post bypass; 3) attending physician signed in upon

transport from operating room; 4)attending signed in during the minute

of the case which yielded the temp used for the measure, or any

additional ideas?

a. Discussion: Abernathy: Whole case? I think most outcomes

should be attributed to person the person who is in the case the

majority of the time. Muehlschlegel: if we can get reliable data

on rewarming post-bypass, we should consider post-bypass.

Abernathy: How do you attribute PONV that relies on timing for

other measures? Buehler: It is attributed to all providers signed

into the case for ≥ 40 minutes. We assume they will have

discussions and determine that someone is responsible.

Abernathy: Is that the answer here? If you started the cases and

you were there for two-hours. Meridith via chat: All providers

for a given case who are signed in ≥ 40 minutes. If a given case

is ≤ 60 minutes, all providers are responsible. Janda: I suggest

we take a hybrid approach and use the starting time of

“conclusion of bypass” since identifying the time rewarming

started during bypass may be challenging, but we would be able

to see timing of “bypass end” within a few minutes more

reliably. Katta: we are all in agreement that merely using the

attending at the time the temperature is taken is not

acceptable. A hybrid approach is nice to include all those in the

case after bypass should be attributed. Janda: Also, as an aside,

this is not meant to be punitive, it is meant to be constructive

feedback with the end goal to have all parties that could

improve to take an extra look at the case. I would hedge on

overinclusion to draw the attention of more people to ensure

providers have what they need. Shook: I like including everyone

and would vote for post bypass. Mathis: would you include pre-

bypass? Shook: No, I would start at bypass, I do not think pre-

bypass matters. Katta: For simplicity should we just grab

everyone signed into the case to start? Janda, we don’t want to

err in the other direction and provide too much information.

We don’t want to frustrate providers by giving them feedback

on something they couldn’t intervene on. I like the idea of

narrowing the window to incorporate those during the post-

bypass period.

b. Janda: Thanks to everyone for their feedback, we have a

starting point and will send out information after meeting

including a drafted measure specification.

d. Hypothermia Avoidance Next Steps

i. We will incorporate your feedback and draft a measure specification

ii. Apply the measure specification to past cases and test functionality

iii. Circulate the measure specification for approval

iv. Synergize efforts with SCA Quality & Safety Committee / CPI Subcommittees

1. Should we engage with SCA Quality & Safety or CPI Committees: Shook:

Involves another layer, their job is to create things that we do not have

to synthesize anymore because they do the job. Does MPOG want to

create a synergized effort and cross relationship? Personally, I’m for

inclusivity to make sure we do the best. There will be a little bit of

politics but those are little humps that shouldn’t get in the way.

2. Mathis: From a broad QA framework, we can include them broadly in a

measure specific way. Its’ better to have the conversation than to stay

in our silos.

4. 2020-2021 Plans

a. Build 1 cardiac-specific measure in 2020/early 2021

i. Post-bypass hypothermia avoidance

b. Build 2-3 cardiac-specific measures in 2021

i. Glucose management?

c. On-bypass hyperthermia avoidance?

i. Antibiotic timing?

d. More discussion

i. Hypotension avoidance

ii. AKI avoidance

e. Opportunities for STS-merged outcome reports requires institutions to integrate with

STS – see surgical registry page on MPOG website

5. New Cardiac- Specific Reporting Dashboard released in beta version

a. See slides for screenshot of reporting dashboard

b. Visit the mpog.org website and click on the blue login button in the top right corner of

the website to login and view: for access issues- please contact:

ajanda@med.umich.edu

6. Subcommittee membership and meeting schedule

a. Open to all anesthesiologists or those interested in improving cardiothoracic measures

i. Do not have to practice at an active MPOG institution

b. Proposed 2021/2022 Meeting Schedule

i. Spring 2021 Meeting: April/May, 2021

ii. Summer 2021 Meeting: July/August, 2021

iii. Fall 2021 Meeting: October/November, 2021

iv. Winter, 2022 Meeting: January/February, 2022

Meeting adjourned at 1:00pm